Machine-learning nonconservative dynamics for new-physics detection
Name
PhysRevE.104.055302.pdf
Description
Published version
Size
1.66 MB
Format
Adobe PDF
Checksum (MD5)
cd5d4817ca9f51cbb746db4b77ee494f
Author(s)
Liu, Ziming
Wang, Bohan
Meng, Qi
Chen, Wei
Tegmark, Max
Liu, Tie-Yan
Date Issued
2021
Journal
Physical Review E
Publisher
American Physical Society (APS)
Citation
Liu, Ziming, Wang, Bohan, Meng, Qi, Chen, Wei, Tegmark, Max et al. 2021. "Machine-learning nonconservative dynamics for new-physics detection." Physical Review E, 104 (5).
Version
Final published version
Abstract
Energy conservation is a basic physics principle, the breakdown of which often implies new physics. This paper presents a method for data-driven "new physics" discovery. Specifically, given a trajectory governed by unknown forces, our neural new-physics detector (NNPhD) aims to detect new physics by decomposing the force field into conservative and nonconservative components, which are represented by a Lagrangian neural network (LNN) and an unconstrained neural network, respectively, trained to minimize the force recovery error plus a constant λ times the magnitude of the predicted nonconservative force. We show that a phase transition occurs at λ=1, universally for arbitrary forces. We demonstrate that NNPhD successfully discovers new physics in toy numerical experiments, rediscovering friction (1493) from a damped double pendulum, Neptune from Uranus' orbit (1846), and gravitational waves (2017) from an inspiraling orbit. We also show how NNPhD coupled with an integrator outperforms both an LNN and an unconstrained neural network for predicting the future of a damped double pendulum.
MIT Department
Massachusetts Institute of Technology. Department of Physics
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1103/PHYSREVE.104.055302
